Ensuring useful adoption of generative artificial intelligence in healthcare

Journal of the American Medical Informatics Association, Journal Year: 2024, Volume and Issue: 31(6), P. 1441 - 1444

Published: March 7, 2024

Abstract Objectives This article aims to examine how generative artificial intelligence (AI) can be adopted with the most value in health systems, response Executive Order on AI. Materials and Methods We reviewed technology has historically been deployed healthcare, evaluated recent examples of deployments both traditional AI (GenAI) a lens value. Results Traditional GenAI are different technologies terms their capability modes current deployment, which have implications systems. Discussion when applied framework top-down realize healthcare. short term unclear value, but encouraging more bottom-up adoption potential provide benefit systems patients. Conclusion healthcare for patients adapt culturally grow this new its patterns.

Language: Английский

---

ChatGPT vs. neurologists: a cross-sectional study investigating preference, satisfaction ratings and perceived empathy in responses among people living with multiple sclerosis

Journal of Neurology, Journal Year: 2024, Volume and Issue: 271(7), P. 4057 - 4066

Published: April 3, 2024

Abstract Background ChatGPT is an open-source natural language processing software that replies to users’ queries. We conducted a cross-sectional study assess people living with Multiple Sclerosis’ (PwMS) preferences, satisfaction, and empathy toward two alternate responses four frequently-asked questions, one authored by group of neurologists, the other ChatGPT. Methods An online form was sent through digital communication platforms. PwMS were blind author each response asked express their preference for questions. The overall satisfaction assessed using Likert scale (1–5); Consultation Relational Empathy employed perceived empathy. Results included 1133 (age, 45.26 ± 11.50 years; females, 68.49%). ChatGPT’s showed significantly higher scores (Coeff = 1.38; 95% CI 0.65, 2.11; p > z < 0.01), when compared neurologists’ responses. No association found between ChatGPT’ mean 0.03; − 0.01, 0.07; 0.157). College graduate, high school education responder, had lower likelihood prefer (IRR 0.87; 0.79, 0.95; 0.01). Conclusions ChatGPT-authored provided than neurologists. Although AI holds potential, physicians should prepare interact increasingly digitized patients guide them on responsible use. Future development consider tailoring AIs’ individual characteristics. Within progressive digitalization population, could emerge as helpful support in healthcare management rather alternative.

Language: Английский

---

Integration of AI in healthcare requires an interoperable digital data ecosystem

Nature Medicine, Journal Year: 2024, Volume and Issue: 30(3), P. 631 - 634

Published: Jan. 30, 2024

Language: Английский

---

Performance of Large Language Models on Medical Oncology Examination Questions

JAMA Network Open, Journal Year: 2024, Volume and Issue: 7(6), P. e2417641 - e2417641

Published: June 18, 2024

Importance Large language models (LLMs) recently developed an unprecedented ability to answer questions. Studies of LLMs from other fields may not generalize medical oncology, a high-stakes clinical setting requiring rapid integration new information. Objective To evaluate the accuracy and safety LLM answers on oncology examination Design, Setting, Participants This cross-sectional study was conducted between May 28 October 11, 2023. The American Society Clinical Oncology (ASCO) Self-Assessment Series ASCO Connection, European Medical (ESMO) Examination Trial questions, original set board-style multiple-choice questions were presented 8 LLMs. Main Outcomes Measures primary outcome percentage correct answers. oncologists evaluated explanations provided by best for accuracy, classified types errors, estimated likelihood extent potential harm. Results Proprietary 2 correctly answered 125 147 (85.0%; 95% CI, 78.2%-90.4%; P &amp;lt; .001 vs random answering). outperformed earlier version, proprietary 1, which 89 (60.5%; 52.2%-68.5%; .001), open-source LLM, Mixtral-8x7B-v0.1, 87 (59.2%; 50.0%-66.4%; .001). contained no or minor errors 138 (93.9%; 88.7%-97.2%). Incorrect responses most commonly associated with in information retrieval, particularly recent publications, followed erroneous reasoning reading comprehension. If acted upon practice, 18 22 incorrect (81.8%; 59.7%-94.8%) would have medium high moderate severe Conclusions Relevance In this performance remarkable performance, although raised concerns. These results demonstrated opportunity develop improve health care clinician experiences patient care, considering impact capabilities safety.

Language: Английский

---

Generative artificial intelligence in surgery

Surgery, Journal Year: 2024, Volume and Issue: 175(6), P. 1496 - 1502

Published: April 6, 2024

Language: Английский

---

An evaluation framework for clinical use of large language models in patient interaction tasks

Nature Medicine, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 2, 2025

Language: Английский

---

Open-Source Large Language Models in Radiology: A Review and Tutorial for Practical Research and Clinical Deployment

Radiology, Journal Year: 2025, Volume and Issue: 314(1)

Published: Jan. 1, 2025

Open-source large language models and multimodal foundation offer several practical advantages for clinical research objectives in radiology over their proprietary counterparts but require further validation before widespread adoption.

Language: Английский

---

DeepSeek in Healthcare: Revealing Opportunities and Steering Challenges of a New Open-Source Artificial Intelligence Frontier

Cureus, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 18, 2025

Generative Artificial Intelligence (GAI) has driven several advancements in healthcare, with large language models (LLMs) such as OpenAI's ChatGPT, Google's Gemini, and Microsoft's Copilot demonstrating potential clinical decision support, medical education, research acceleration. However, their closed-source architecture, high computational costs, limited adaptability to specialized contexts remained key barriers universal adoption. Now, the rise of DeepSeek's DeepThink (R1), an open-source LLM, gaining prominence since mid-January 2025, new opportunities challenges emerge for healthcare integration AI-driven research. Unlike proprietary models, DeepSeek fosters continuous learning by leveraging publicly available datasets, possibly enhancing ever-evolving knowledge scientific reasoning. Its transparent, community-driven approach may enable greater customization, regional specialization, collaboration among data researchers clinicians. Additionally, supports offline deployment, addressing some privacy concerns. Despite these promising advantages, presents ethical regulatory challenges. Users' worries have emerged, concerns about user retention policies developer access user-generated content without opt-out options. when used applications, its compliance China's data-sharing regulations highlights urgent need clear international governance. Furthermore, like other LLMs, face limitations related inherent biases, hallucinations, output reliability, which warrants rigorous validation human oversight before application. This editorial explores role workflows, while also highlighting security, accuracy, responsible AI With careful implementation, considerations, collaboration, similar LLMs could enhance innovation, providing cost-effective, scalable solutions ensuring expertise remains at forefront patient care.

Language: Английский

---

Large language models improve the identification of emergency department visits for symptomatic kidney stones

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: Jan. 28, 2025

Recent advancements of large language models (LLMs) like generative pre-trained transformer 4 (GPT-4) have generated significant interest among the scientific community. Yet, potential these to be utilized in clinical settings remains largely unexplored. In this study, we investigated abilities multiple LLMs and traditional machine learning analyze emergency department (ED) reports determine if corresponding visits were due symptomatic kidney stones. Leveraging a dataset manually annotated ED reports, developed strategies enhance including prompt optimization, zero- few-shot prompting, fine-tuning, augmentation. Further, implemented fairness assessment bias mitigation methods investigate disparities by with respect race gender. A expert assessed explanations GPT-4 for its predictions they sound, factually correct, unrelated input prompt, or potentially harmful. The best results achieved (macro-F1 = 0.833, 95% confidence interval [CI] 0.826–0.841) GPT-3.5 0.796, CI 0.796–0.796). Ablation studies revealed that initial model benefits from fine-tuning. Adding demographic information prior disease history prompts allows make better decisions. Bias found exhibited no racial gender disparities, contrast GPT-3.5, which failed effectively diversity.

Language: Английский

---

A Bibliometric Analysis of the Rise of ChatGPT in Medical Research

Medical Sciences, Journal Year: 2023, Volume and Issue: 11(3), P. 61 - 61

Published: Sept. 17, 2023

The rapid emergence of publicly accessible artificial intelligence platforms such as large language models (LLMs) has led to an equally increase in articles exploring their potential benefits and risks. We performed a bibliometric analysis ChatGPT literature medicine science better understand publication trends knowledge gaps. Following title, abstract, keyword searches PubMed, Embase, Scopus, Web Science databases for published the medical field, were screened inclusion exclusion criteria. Data extracted from included articles, with citation counts obtained PubMed journal metrics Clarivate Journal Citation Reports. After screening, 267 study, most which editorials or correspondence average 7.5 +/- 18.4 citations per publication. Published on authored largely United States, India, China. topics discussed use accuracy research, education, patient counseling. Among non-surgical specialties, radiology ChatGPT-related while plastic surgery among surgical specialties. number top 20 most-cited was 60.1 35.3. journals publications, there 10 3.7 publications. Our results suggest that managing inevitable ethical safety issues arise implementation LLMs will require further research capabilities ChatGPT, generate policies guiding adoption science.

Language: Английский

---